# Unusual Upvote Trends (Discussion)

Recently, I answered a question, How exactly does gravity work?, which received as of now, 30 upvotes, and I am glad that the community found the answer helpful, and of above average quality. However, many of my other answers, in my opinion, have been much better in both content, length, style, approach, etc. Yet, those received a few votes, whilst the aforementioned a phenomenal 30!

Hence, I ask, does anyone have any theories or input they'd like to share regarding what they think serial upvoted answers have that make them distinct from others? Or perhaps is it less deterministic?

I'm just puzzled as to how I got 30 votes.

Note: this is in no way a rant about other answers not receiving votes; I'm simply curious. Also, it's certainly not a complaint! :)

• And if you get a popular link in Hacker News, it's easy to get over 100 points. – jinawee Jun 5 '14 at 7:49
• It is not uncommon for a response to get to the top of a question, and then subsequent readers will read the top answer, upvote it, and not read further answers. – Jerry Schirmer Jun 5 '14 at 18:35
• Guys, an interesting observation is on this site you have to 'deal with' three groups: actual working physicists; the "popular science" crowd (like me); and, the "keen and admirable but mysteries-of-entanglement leaning" crowd. (Might as well give them three offensive nicknames .. "the real eggheads", "the almost-finished-that-degree set", and the "ufo nuts!".) The same phenomenon happens on other lists but not as strongly. (You guys just have to suck it up - you're dealing with the mysteries of the universe for the rest of us, for goodness sake :) ) I believe ...cont'd – Fattie Jun 6 '14 at 6:21
• ... I believe this explains some of the voting mysteries you mention. {I wrote a long essay about this phenomenon, and was about to post it here, but decided to delete it since I didn't want to clutter up your site with more "group two, the Mediocres" blather!} Cheers! – Fattie Jun 6 '14 at 6:22
• BTW @JamalS on that specific question. I believe the actual metric at work is: notice the ticked answer is the group 2, popular-science, answer (featuring the inevitable "wildly confusing earth-in-drain with bent white graph lines" diagram :) ). I believe you're getting group-1-counter-votes there: people in group1 are seeing the ticked group2 answer, rolling their eyes and thinking "Thanks for that, Michio Kaku", and ticking your group1 answer as a counterbalance. (That's certainly what I did!) (Many group2 people are group1 wannabes ;-) ) – Fattie Jun 6 '14 at 6:29

To the right of every page is the Hot Network Questions list. If you note the list carefully, you'll find the question you linked to appear on the list:

While not the clearest picture (looks better on my screen even after cropping it), there's your answer and the hot question.

I have three such answers (3 links there) to my name due to the HNQ-effect, and I know I'm not the only one who's seen this.

• Yep, and in particular the question has a very simple title, which many visitors to all SE sites will (think they can) understand, so it will attract a lot more attention than a technical question that makes it on to the hot question list. The lesson to take away is that absolute vote counts mean little about a question's or answer's quality unless they're normalized to the amount of attention the post is getting. – David Z Jun 4 '14 at 21:49
• This is very likely the cause here, but in truth even before the Hot Questions sidebar most long time users agreed that their highest voted questions and the content they were most proud of were disjoint sets. Very high vote counts usually indicate an accessible answer to an accessible question rather than indicating some very deep pontifications. – dmckee --- ex-moderator kitten Jun 4 '14 at 21:56

That's nothing! I got 24 votes for a picture of a fly's foot :-)

I suspect that the vast majority of the site members are interested amateurs rather than physics students or people being paid to do physics. As an (ex) scientist my favourite answers are the ones that are right at the edge of what I know, but these are exactly the answers that are going to be uninteresting to most other people. My experience is that you get the most votes when you take a question that everyone can understand and provide an answer that everyone can understand.

But I agree that it certainly helps if the question you're answering makes the hot questions list.

• A fly's foot? Brilliant! – JamalS Jun 5 '14 at 11:46
• I agree that easy answers to easy questions seem to yield easy internet points. My highest upvoted answer is another example: I applied $s=v/t$! Hah! – Danu Jun 8 '14 at 2:57
• ... and now you got 11 votes for mentioning a picture of a fly's foot... chuckles – user42659 Jun 18 '14 at 13:42

I'm pretty much rehashing what everyone else has said, but what is the point of a discussion if not to repeat the same few points that everyone agrees with in many different ways?

There are many contributing factors to how many votes an answer might get. The most obvious and influential being whether or not the answer is correct and complete. However, it is far from unheard of that a masterfully made answer with just the right information presented in an easy to understand way might get few or even no votes.

Vote-hunting (that's a new term I just made up) is a skill; sadly one I am not very adept at, but I am an adept pattern recognizer so I see the trends. Beyond being correct and easily understood, it particularly helps if the question has a simple title; one most users could understand and most non-physicists would think "hey, that sounds pretty cool!" That said, too simple and it's likely to be ignored by everyone. The question you gave as an example is perfect for this. "How exactly does gravity work?" That type of title will definitely bring the viewers. The other thing that helps is if a question has a lot of up votes. That tells users that most people like the question and find it interesting. This makes them more interested in seeing what the answer to it is.

Once they've clicked in, you want the viewer to stay. To that end, the question needs to be well worded, easily followable, and interesting to novices and at least undergrad level as well. You might consider editing the format or grammar of the question if you feel it needs improvement in those areas but remember do not make any substantive changes to the content of the question unless it is your own. Also, consider improving the tags to include not only subjects mentioned in the question, but also subjects you know would have to be mentioned in any correct answer.

Medium to long answers usually are more correct than short answers and so they tend to be read first and upvoted. But too long and nobody wants to take the time. This is also true with math-heavy answers, answers that aren't segmented with short, manageable paragraphs, or answer that start with things like "This is just my opinion, I really don't know for sure..." Sometimes, a lot of math is called for, so then it's fine.

Furthermore, try to make it fun to read. Put a little piece of unexpected humour right in the middle of your answer, use wild and/or silly scenarios for thought experiments, be creative. Also, images. Let me restate that. Images. When you have a nice image with a splash of colour, that makes people want to read it. If you can find an image that could pretty much answer the question by itself and you pair it up with a nice written explanation that adds additional info for the more curious readers, then you're gold. Answers like that might win you the populist badge.

Avoid the use of this:

Edit:

It's all well and good for telling people that you fixed a problem or added in missing info, but it's exhausting for a reader. It may get you points from a few expert users, but novices rarely recognize the original need for the edit and I've seen it happen many times where the votes abruptly stop coming in once someone has added the edit line.